library(ggplot2)
library(dplyr)
library(RColorBrewer)
library(ambient)
library(purrr)
set.seed(123)Set Up
Art Piece 1
# Parameters
num_walks <- 500
num_steps <- 100
step_size_sd <- 0.1
# Generate random walk data
walk_data <- map_df(1:num_walks, ~{
x_pos <- 0
y_pos <- 0
path_x <- numeric(num_steps)
path_y <- numeric(num_steps)
for (i in 1:num_steps) {
x_pos <- x_pos + rnorm(1, 0, step_size_sd)
y_pos <- y_pos + rnorm(1, 0, step_size_sd)
path_x[i] <- x_pos
path_y[i] <- y_pos
}
tibble(
walk_id = .x,
step_id = 1:num_steps,
x = path_x,
y = path_y
)
})
# Create the plot
ggplot(walk_data, aes(x = x, y = y, group = walk_id, color = factor(walk_id))) +
geom_path(alpha = 0.15, linewidth = 0.3) +
scale_color_manual(values = rep(brewer.pal(11, "Spectral"), length.out = num_walks)) +
theme_void() +
coord_fixed() +
theme(
legend.position = "none", # Hide legend for cleaner look
plot.background = element_rect(fill = "black", color = "black")
)
Title: Stochastic Bloom
Museum-Style Description:
This piece, Stochastic Bloom, captures the inherent tension between order and chaos that governs complex systems. Each luminous trajectory, born from a simple generative rule, embarks on a journey of apparent randomness. Yet, from this multitude of individual paths, a collective, almost organic structure emerges. The artist invites the viewer to contemplate the unseen algorithms that shape our world, from the flocking of birds to the ebb and flow of digital information. It is a meditation on emergence, where simplicity begets profound complexity, and data points transform into ethereal dance.
Description of the Code Choices and How They Influenced the Appearance:
Data Generation: The piece uses data generated from
num_walks2D random walks. Each walk starts from the center of the canvas (0,0) and takesnum_stepssteps. The step for each walk is generated usingrnorm()for both x and y increments with a standard deviation ofstep_size_sd, ensuring a degree of unpredictability in their paths.Plotting with
ggplot2: Theggplot2package is the primary tool.geom_path(): Each random walk is rendered as a separate path. By usinggeom_pathfor each of thenum_walksrandom walks, we trace the journey of each data series as a continuous line, creating the filamentous structure. Thelinewidthis set to0.3for thin lines.Aesthetics & Color:
alpha(transparency): Each path is drawn withalpha = 0.15. Setting a low alpha value for the paths allows their overlaps to create areas of greater density and visual weight, highlighting regions where many paths converge or cross.color: The color of each line is mapped to itswalk_idusingfactor(walk_id). A repeating spectral palette fromRColorBrewer(brewer.pal(11, "Spectral")) is used. By mapping thecoloraesthetic to the unique ID of each random walk using a repeating spectral palette, a vibrant differentiation between lines is achieved, preventing a monotonous visual and adding depth.
Theme and Axes:
theme_void(): This theme is used to remove axes, gridlines, and most background elements. Employingtheme_void()strips the canvas of traditional chart elements, focusing the viewer’s attention entirely on the artistic form of the data itself.plot.background = element_rect(fill = "black", color = "black"): The plot background is set to black to make the colored lines stand out.coord_fixed(): This ensures that the aspect ratio is maintained, so the random walks spread evenly in all directions. Usingcoord_fixed()maintains a 1:1 aspect ratio, preventing distortion and ensuring the ‘explosion’ of lines feels symmetrical and balanced in its spread.legend.position = "none": The legend is hidden to maintain a clean, artistic look.
Art Piece 2
# Parameters for Perlin noise
grid_width <- 500
grid_height <- 300
noise_frequency <- 0.02 # Lower frequency for larger, smoother patterns
noise_octaves <- 5 # More octaves for more detail/richness
# Generate Perlin noise
# Using long_grid for a more convenient format for ggplot
noise_data <- long_grid(
x = seq(0, 1, length.out = grid_width),
y = seq(0, 1, length.out = grid_height)
) %>%
mutate(
noise_value = gen_perlin(x, y, frequency = noise_frequency, octaves = noise_octaves)
)
# Create the plot
ggplot(noise_data, aes(x = x, y = y, fill = noise_value)) +
geom_raster(interpolate = TRUE) + # interpolate = TRUE for smoother transitions
scale_fill_gradientn(
colors = c("#000033", "#000066", "#2E0854", "#4B0082", "#8A2BE2", "#FF00FF", "#FF69B4", "#1E90FF", "#00BFFF", "#AFEEEE"),
name = "" # No legend title
) +
theme_void() +
coord_fixed(expand = FALSE) + # expand = FALSE to fill the entire plot area
theme(
legend.position = "none", # Hide legend for purely artistic view
plot.background = element_rect(fill = "black", color = "black")
)
Title: Rstrology
Museum-Style Description:
Behold Rstrology, a sublime confluence where the rigorous logic of R code meets the enigmatic allure of the cosmos, unfurling as a digital aurora of celestial gradients. Bands of vibrant, otherworldly light—deep cosmic indigos melting into spectral purples, superheated magentas, and ethereal cyans—streak across the canvas, suggesting distant nebulae or the very fabric of a synthetic universe taking form. The artist harnesses R to sculpt new firmaments, inviting contemplation on whether these are the birth colors of stars, the resonant frequencies of data streams, or a horoscope cast in code. Rstrology challenges us to find the poetic in the programmatic, seeking the soul in the machine and the universal patterns that bind the stars to the silicon chip, offering a meditation on the inherent beauty found in the structured randomness governing both galaxies and algorithms.
Description of the Code Choices and How They Influenced the Appearance:
Core Generative Technique: Perlin Noise: The foundation of this piece is 2D Perlin noise generated using the
ambientpackage in R.gen_perlin(): A 2D grid of Perlin noise values is created usingambient::gen_perlin()on a grid generated byambient::long_grid(). The grid dimensions aregrid_widthbygrid_height. By generating a matrix of Perlin noise, we obtain a set of smoothly varying values that form the fundamental structure of the image’s light and dark regions.Frequency/Scale: The
frequencyparameter ingen_perlin()is set tonoise_frequency. Choosing a relatively low frequency for the Perlin noise results in large, sweeping patterns and gentle gradations rather than fine, chaotic detail, contributing to the sense of vastness.Octaves: The
octavesparameter is set tonoise_octaves. Using multiple octaves adds layers of noise at different frequencies, creating a richer and more detailed texture within the larger forms.
Color Mapping with
ggplot2:geom_raster(): The noise matrix is visualized usinggeom_raster. Theinterpolate = TRUEargument helps smooth the color transitions between pixels. Thegeom_raster()function maps each noise value directly to a pixel on the canvas.interpolate = TRUEfurther enhances the smoothness, creating a more fluid visual field.scale_fill_gradientn(): A custom multi-color gradient is used to map noise values to colors. The chosen colors arec("#000033", "#000066", "#2E0854", "#4B0082", "#8A2BE2", "#FF00FF", "#FF69B4", "#1E90FF", "#00BFFF", "#AFEEEE"). By applying this carefully selected multi-color gradient—transitioning from deep blues and purples through vibrant pinks to bright cyans—the raw noise values are transformed into the otherworldly hues that define the piece’s atmosphere.
Canvas Styling:
theme_void(): This removes all axes, gridlines, and background elements. The use oftheme_void()removes all distracting chart elements, immersing the viewer fully in the generated visual.coord_fixed(expand = FALSE): This maintains the aspect ratio of the noise grid and ensures the raster fills the entire plot area without padding. Usingcoord_fixed(expand = FALSE)ensures the noise patterns are not unnaturally stretched and that the artwork extends to the very edges of the frame.legend.position = "none": The color scale legend is hidden.plot.background = element_rect(fill = "black", color = "black"): The plot background is set to black to enhance the luminosity of the colors.